Stats 253: Analysis of Spatial and Temporal Data

Description

A unified treatment of methods for spatial data, time series, and other correlated data from the perspective of regression with correlated errors. Two main paradigms for dealing with autocorrelation: covariance modeling (kriging) and autoregressive processes. Bayesian methods. Computational issues will be a focus of this class.

Prerequisites: statistical inference (STATS 200) and linear regression with linear algebra (STATS 203). Alternatively, if you've taken CS 229, you should be fine.

Meetings

This course is divided into two halves:

For the first 4 weeks, the class will meet Mondays, Wednesdays, and Fridays, 2:15-3:30 PM in TBA. This is so we can cover enough material for you to begin working on a project.
For the last 4 weeks, classes will meet only sporadically to give you time to work on your projects. Lectures will cover special topics that are requested by students in the class.

Instructors

Dennis Sun, Sequoia 238. Office hours: After class or by appointment.
Jingshu Wang, Sequoia 233. Office hours: Thursdays 2-4pm.

Please contact us through the Piazza forum. This helps us stay organized. There is an option to send private messages to instructors through Piazza as well.

Requirements

3 short quizzes. These will be straightforward and are intended just to help you keep up with the material and to help me gauge the class's understanding. These will be graded check / resubmit.
3 data analysis assignments. These will be graded check / resubmit.
Class project (optional for those taking the course CR/NC).

References

There is no natural textbook for this class. I will supply lecture slides and notes when necessary. However, the following references may be helpful.

A. Agresti. Foundations of Linear and Generalized Linear Models. Wiley 2015.

R. S. Bivand et al. Applied Spatial Data Analysis with R. [access online] 2nd edition. Springer 2013.

N. Cressie. Statistics for Spatial Data. Revised edition. Wiley 1993.

S. Banerjee, B. P. Carlin, and A. E. Gelfand. Hierarchical Modeling and Analysis for Spatial Data. [access online] Chapman and Hall 2003.